Background: We investigated whether opportunistic screening for osteoporosis can be done from computed tomography (CT) scans of the wrist/forearm using machine learning. Methods: A retrospective study of 196 patients aged 50 years or greater who underwent CT scans of the wrist/forearm and dual-energy X-ray absorptiometry (DEXA) scans within 12 months of each other was performed. Volumetric segmentation of the forearm, carpal, and metacarpal bones was performed to obtain the mean CT attenuation of each bone. The correlations of the CT attenuations of each of the wrist/forearm bones and their correlations to the DEXA measurements were calculated. The study was divided into training/validation (n = 96) and test (n = 100) datasets. The performance of multivariable support vector machines (SVMs) was evaluated in the test dataset and compared to the CT attenuation of the distal third of the radial shaft (radius 33%). Results: There were positive correlations between each of the CT attenuations of the wrist/forearm bones, and with DEXA measurements. A threshold hamate CT attenuation of 170.2 Hounsfield units had a sensitivity of 69.2% and a specificity of 77.1% for identifying patients with osteoporosis. The radial-basis-function (RBF) kernel SVM (AUC = 0.818) was the best for predicting osteoporosis with a higher AUC than other models and better than the radius 33% (AUC = 0.576) (p = 0.020). Conclusions: Opportunistic screening for osteoporosis could be performed using CT scans of the wrist/forearm. Multivariable machine learning techniques, such as SVM with RBF kernels, that use data from multiple bones were more accurate than using the CT attenuation of a single bone.

Purpose: Transarterial radioembolization can serve as an ablative therapy for early-stage hepatocellular carcinoma (HCC). Given the volumetric variability of liver segments, this study aimed to characterize the safety of ablative radioembolization by determining percent liver treated (%LT) thresholds associated with biochemical toxicity. Patients and Methods: Patients with HCC receiving a single ablative radioembolization treatment using glass microspheres from 2017 through 2020 were reviewed. %LT was calculated as treatment angiosome volume divided by whole liver volume. Biochemical toxicities were defined as increases in Albumin-Bilirubin (ALBI) grade or Child-Pugh (CP) class compared to baseline and albumin or bilirubin adverse events (AEs) per the Common Terminology Criteria for Adverse Events. Receiver operating characteristic curves and multivariate logistic regression analyses were performed to assess the impact of %LT on toxicities. Results: Of 141 patients analyzed, 53% (n=75) were ALBI 1, 45% (n=64) ALBI 2, 79% (n=111) CP-A, and 21% (n=30) CP-B. A %LT ≥ 14.5% was associated with grade/class increases in ALBI 2 (p≤ 0.01) and CP-B patients (p=0.026). In multivariate analysis, a %LT ≥ 14.5% was an independent predictor of increases in the ALBI 2 and CP-B groups (p< 0.01). No significant %LT threshold was found for ALBI 1 and CP-A patients. No grade 3/4 albumin or bilirubin AEs were reported, while grade 2 AEs were related to an initial whole liver volume < 1.3 L (p≤ 0.01). Conclusion: Patients with ALBI 2 and CP-B liver function are less likely to have an increase in their respective grade/class when treating < 14.5% of the liver using glass microspheres. ALBI 1 and CP-A patients showed no definitive %LT threshold for biochemical toxicity within the range of this study.

BACKGROUND AND PURPOSE: Parry-Romberg syndrome is a rare disorder characterized by progressive hemifacial atrophy. Concomitant brain abnormalities have been reported, frequently resulting in epilepsy, but the frequency and spectrum of brain involvement are not well-established. This study aimed to characterize brain abnormalities in Parry-Romberg syndrome and their association with epilepsy. MATERIALS AND METHODS: This is a single-center, retrospective review of patients with a clinical diagnosis of Parry-Romberg syndrome and brain MR imaging. The degree of unilateral hemispheric atrophy, white matter disease, microhemorrhage, and leptomeningeal enhancement was graded as none, mild, moderate, or severe. Other abnormalities were qualitatively reported. Findings were considered potentially Parry-Romberg syndrome–related when occurring asymmetrically on the side affected by Parry-Romberg syndrome. RESULTS: Of 80 patients, 48 (60%) had brain abnormalities identified on MR imaging, with 26 (32%) having abnormalities localized to the side of the hemifacial atrophy. Sixteen (20%) had epilepsy. MR imaging brain abnormalities were more common in the epilepsy group (100% versus 48%, P < .001) and were more frequently present ipsilateral to the hemifacial atrophy in patients with epilepsy (81% versus 20%, P < .001). Asymmetric white matter disease was the predominant finding in patients with (88%) and without (23%) epilepsy. White matter disease and hemispheric atrophy had a higher frequency and severity in patients with epilepsy (P < .001). Microhemorrhage was also more frequent in the epilepsy group (P = .015). CONCLUSIONS: Ipsilateral MR imaging brain abnormalities are common in patients with Parry-Romberg syndrome, with a higher frequency and greater severity in those with epilepsy. The most common findings in both groups are white matter disease and hemispheric atrophy, both presenting with greater severity in patients with epilepsy.